Curved Grade Calculator

Start by entering your students' test scores. You can paste a comma separated list of numbers into the input field, or type each score individually and press Enter to add it. If you want to experiment first, click the "Sample class of 20" button to load a realistic set of scores.

Next, select one or more curving methods using the checkboxes. Each method will appear as its own results card showing the curved statistics, grade distribution, and a bar chart comparing original grades to curved grades.

Adjust the target mean (default 75) to match your goal. The calculator will recommend the method that brings the class average closest to that target. You can also see how many students moved from failing to passing under each curve.

Flat curve adds the same number of points to every score. It is the simplest method and preserves the exact spacing between students. If you add 5 points, every score moves up by exactly 5. The downside is that students who already scored high may exceed 100.

Square root curve takes the square root of the original score and multiplies by 10. This compresses the range, boosting lower scores more than higher ones. A student who scored 49 would receive a 70, while a student who scored 81 would receive a 90. It is useful when the test was genuinely too difficult for most students.

Scale to highest makes the top score equal to 100 and scales everyone else proportionally. This works well when no one achieved a perfect score but at least one student came close. If the highest score is already very low, the scaling factor may push everyone up dramatically.

Bell curve assigns letter grades based on standard deviation from the mean. The top 10% receive an A, the next 20% a B, the middle 40% a C, the next 20% a D, and the bottom 10% an F. This guarantees a fixed distribution regardless of raw scores.

Drop lowest removes the lowest N scores from the set before computing the class average. This is most commonly used across multiple assignments rather than within a single test, but it can illustrate the effect of removing outlier scores on the overall distribution.

Curving is appropriate when external factors made a test harder than intended. If the class average on an exam is significantly lower than expected and the distribution suggests most students struggled with the same material, a curve can correct for test difficulty without penalizing students for an instructional gap.

On the other hand, curving should not be used routinely to compensate for lack of preparation. If students consistently need curves to pass, the issue may lie with the instruction, the pacing, or the alignment between the material taught and the material tested. Addressing the root cause is more effective than applying a curve after every exam.

It is also worth considering the size of the class. Curves based on relative performance (like the bell curve) work better with larger groups of 30 or more students, where statistical distributions are more stable. In a small class, one or two outliers can distort the results significantly.

Different curving methods change the grade distribution in different ways. A flat curve shifts the entire distribution to the right without changing its shape. The standard deviation, range, and relative rankings all stay the same. Only the mean moves.

The square root curve compresses the distribution, pulling low scores up more aggressively than high scores. This reduces the standard deviation and can eliminate extreme gaps between the strongest and weakest students. It tends to produce more B and C grades while reducing the number of F grades.

Scaling to the highest score stretches the distribution so that the top score becomes 100. If the highest score was already close to 100, the effect is minimal. If the highest score was low (say 72), every score gets multiplied by a large factor, which can create an unrealistically compressed distribution near the top.

The bell curve forces a predetermined distribution regardless of actual performance. This means that even if every student scored above 90, some would still receive a D or F. This method is controversial for this reason and is most commonly used in large university courses where normalization across sections is needed.